This may not be a feature request for the Impulse Studio but It would be nice to have a category in this forum dedicated to discussions regarding the core of Deep Learning like architecture/model choice, training (loss, optimization, regularization etc.), evaluation metrics, model improvement and deployment. This could also be a space where the community members can share their latest exciting developments and tips on what pre-processing, feature extraction works best for certain types of data and how to improve and/or adapt existing model for new cases. I really think this will be a great category to have. What do you think?
I think that’s actually a good idea, let me loop internally with the DevRel team to see if we all want to add additional categories (Machine Learning, DSP, Embedded, or other) and which ones.
Thanks for that request,